光电子技术, 2018, 38 (2): 111, 网络出版: 2019-01-18
基于压缩感知的图像配准算法研究
Research on Image Registration Algorithm Based on Compressed Sensing
摘要
针对SIFT算法的图像配准耗时长的问题, 提出一种基于压缩感知的图像配准算法。首先, 使用FAST算法提取图像角点, 然后, 使用SIFT算法生成角点描述子并使用压缩感知原理降维, 最后, 使用双向匹配算法进行匹配。通过实验数据分析, 基于压缩感知的图像配准算法在图像配准准确度方面与传统SIFT算法、传统SURF算法、文献
Abstract
Aiming at the time consuming problem of image registration of SIFT algorithm, an image registration algorithm based on compressed sensing was proposed. Firstly, the image corner was extracted by FAST algorithm, then, the SIFT algorithm was used to generate the corner descriptor and the compressed sensing principle was used to reduce the dimension of the corner descriptor. Finally, a bidirectional matching algorithm was used to match. Through the experimental data analysis, image registration algorithm based on compressed sensing is nearly the same in image registration accuracy in comparison with the traditional SIFT algorithm、the traditional SURF algorithm、reference
章盛, 李培华, 张骏, 钱名思, 鲁兴平, 叶程广. 基于压缩感知的图像配准算法研究[J]. 光电子技术, 2018, 38(2): 111. ZHANG Sheng, LI Peihua, ZHANG Jun, QIAN Mingsi, LU Xingping, YE Chengguang. Research on Image Registration Algorithm Based on Compressed Sensing[J]. Optoelectronic Technology, 2018, 38(2): 111.